Determinants of the export channel selection and export performance

(1)

Durham E-Theses

Determinants of the export channel selection and export

performance

LI, MIN

How to cite:

LI, MIN (2018) Determinants of the export channel selection and export performance, Durham theses, Durham University. Available at Durham E-Theses Online: http://etheses.dur.ac.uk/12571/

Use policy

The full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission or charge, for personal research or study, educational, or not-for-prot purposes provided that:

• a full bibliographic reference is made to the original source • alinkis made to the metadata record in Durham E-Theses • the full-text is not changed in any way

The full-text must not be sold in any format or medium without the formal permission of the copyright holders. Please consult thefull Durham E-Theses policyfor further details.

Academic Support Oce, Durham University, University Oce, Old Elvet, Durham DH1 3HP e-mail: [email protected] Tel: +44 0191 334 6107

(2)

The determinants of the export channel

selection and export performance

Min Li

A thesis submitted in fulfilment of the requirements for

the degree of Doctor of Philosophy in Marketing

Durham University Business School

Durham University

(3)

i Abstract

The aim of this thesis is to explore how firms can make better use of channel selections to create value from their organisational capabilities, such as product development capabilities (PDC), in export operations by considering the role of resource-based factors such as entrepreneurial orientation (EO) and institutional factors such as cultural-cognitive institutional distance (CCID). Through a

systematic review of the contemporary empirical studies on export channel selection, it is found that although previous studies have made advancements in improving our understanding of export channel selection using a variety of theoretical bases, this domain is still immature, in that significant theoretical and methodological gaps exist. Based on this review, the author then carries out an empirical study to explore PDC as the determinant of export channel selection and export performance to address the gaps in current export channel research, integrating resource-based view and institutional theory. Using the data collected from multiple respondents of 294 Chinese export firms, the study finds that firms with higher levels of PDC are more likely to select the hierarchical channel. In addition, the possession of EO negatively moderates the propensity of high PDC-firms to select the hierarchical channel. The study has also found that the moderating effect of EO on PDC-channel selection becomes stronger when the CCID between the home and export market increases, and that the alignment between PDC, EO, CCID, and channel selection can help firms to achieve better export performance. This thesis contributes to the literature by extending the application of RBV and institutional theory in export channel selection, and adds knowledge to the roles of PDC, EO, and CCID in helping firms to achieve better performance in export markets by means of export channel selection.

(4)

ii Publications

Li, Min, Xinming He, and Carlos M. P. Sousa (2017), "A review of the empirical research on export channel selection between 1979 and 2015," International Business Review, 26 (2), 303-23.

(5)

iii Table of Contents

Abstract ... i

Publications ... ii

List of Tables ... v

Declaration and Statement of Copyright ... vi

Acknowledgements ... vii

Chapter 1. Introduction ... 1

Chapter 2. A Review of the Empirical Research on Export Channel Selection between 1979 and 2015 ... 12

2.1. Scope and analytical approach of the review ... 12

2.2. Theoretical bases and frameworks ... 27

2.2.1. Typology of export channel... 28

2.2.2. Theories and frameworks ... 34

2.3. Methodological characteristics of the studies reviewed ... 45

2.3.1. Fieldwork characteristics ... 51

2.3.2. Sampling and data collection ... 52

2.3.3. Statistical analysis ... 55

2.4. Directions and implications ... 55

2.4.1. Theoretical issues ... 56

2.4.2. Methodological issues ... 62

2.4.3. Practical implications ... 65

2.4.4. Directions for new ideas in export channel selection research ... 67

Chapter 3. Product Development Capabilities-based Export Channel Selection and Export performance ... 71

3.1. Introduction ... 71

3.2. Literature review ... 76

3.2.1. RBV and institutional theory ... 79

(6)

iv

3.3.1. Product development capabilities and export channel selection ... 83

3.3.2. The moderating role of entrepreneurial orientation ... 86

3.3.3. The impact of cultural-cognitive institutional distance on the moderating effect of entrepreneurial orientation on PDC-channel relationship ... 89

3.3.4. Capabilities, cultural-cognitive institutional distance, and export performance... 93

3.4. Methodology ... 96

3.4.1. Sampling and data collection ... 96

3.4.2. Measures ... 100

3.4.3. Common method variance ... 106

3.4.4. Construct reliability and validity ... 107

3.5. Empirical findings ... 108

3.5.1. Export channel selection results ... 110

3.5.2. Export performance results ... 110

Chapter 4. Discussion and conclusions ... 115

4.1. Theoretical implications ... 116

4.2. Managerial implications ... 121

4.3. Limitations ... 123

4.4. Conclusions ... 125

(7)

v List of Tables

Table 1. Empirical findings of the study reviewed __________________________ 14 Table 2. Typology and definition of channel structures ______________________ 30 Table 3. Characteristics of the studies reviewed ____________________________ 46 Table 4. Profile of sample ventures ______________________________________ 99 Table 5. Multi-item measures and validity assessment ______________________ 103 Table 6. Descriptive statistics and correlation matrix _______________________ 109 Table 7. Logistic Regression Analysis of Export Channel Selection ___________ 112 Table 8. Regression Analysis of Export Performance _______________________ 113

(8)

vi Declaration and Statement of Copyright

I hereby declare that this PhD thesis entitled “The determinants of the export channel selection and export performance” has been carried out by myself for the degree of Doctor of Philosophy in English under the guidance and supervision of Dr Xinming He and Professor Carlos M.P. Sousa, Durham University Business School, Durham University, UK.

Part of the thesis is published in International Business Review (Li, M. and He, X. and Sousa, C.M.P. (2017) 'A review of the empirical research on export channel selection between 1979 and 2015.', International Business Review., 26 (2). pp. 303-323.).

For the present thesis, which I am submitting to the University, no degree or diploma or distinction has been conferred on me before, either in this or in any other

University.

The copyright of this thesis rests with the author. No quotation from it should be published without the author’s prior written consent and information derived from it should be acknowledged.

Place: Durham Min Li

(9)

vii Acknowledgements

I would like to begin by giving special thanks to my supervisor Dr Xinming He and co-supervisor Professor Carlos M.P. Sousa, for their guidance, support and

encouragement in my research and life. They have kindly led me to research and taught me a great deal, with the utmost patience. Without my supervisors, this thesis would not have been possible, and their strong work ethic and principles will be a continuous inspiration for me in the future.

Secondly, I am extremely grateful to my parents. You have devoted yourselves to ensuring that I was given the best education possible and a promising future. You are my most reliable allies, and the ultimate reason why I feel I can overcome all

difficulties.

I would also like to thank Professor Rob Dixon for awarding me a Dean’s

Scholarship, which provided me with the financial support necessary to study a PhD in marketing. I would also like to thank Professor Susan Hart for sharing her rich experiences in academic research and thank Dr Karena Yan and Dr Cheng Wang for reviewing my annual progress and providing me with great comments and

suggestions about my thesis and research.

In addition, I would like to extend my thanks to the great scholars at Xiamen University, Dr Qun Tan, Professor Shuijun Peng, Dr Shoude Chen, Mrs Qing Liao and Mr Xiaobo Liao, for providing me with networks and convenient access to data collection. Furthermore, I would like to thank all the firms that participated in the survey to enable the data collection of this thesis.

(10)

viii

Lastly, I would like to express my gratitude to all the friends I have met in Durham. I am extremely grateful for your support and friendship; Durham and its people will always be among my most beautiful memories.

(11)

1 Chapter 1. Introduction

Exporting is one of the most important internationalisation strategies by which firms can expand their market base into the international arena, and thereby gain more opportunities and achieve better performance (e.g., Aulakh et al., 1997; He et al., 2013; Klein et al., 1990b). According to the World Bank (2015), exports accounted for around 30% of the global GDP in 2015. In the field of exporting, channel selection is a key strategic decision, being the organisational structure that a

company employs to arrange and support the marketing, selling, and distribution of its products into foreign markets (Anderson et al., 1987a; Hoppner et al., 2015; Klein et al., 1990b).

Widely recognised as one of the most important decisions in a firm’s international marketing strategy, export channel selection has significant cost and performance implications for exporting organisations (Barney et al., 2001; He et al., 2013); for instance, an export channel cannot be easily reversed when chosen and implemented, due to the high level of sunk costs involved (Anderson et al., 1987a; Ramaseshan et al., 1994). An export channel also plays an important role in affecting firms’ export performance, which can exert a vital influence on firms’ willingness to hold and continue their investment and involvement in the foreign market (Brouthers et al., 2008a; Sousa et al., 2008; Zou et al., 1998).

The study of export channels can be traced back to the 1970s when some scholars employed case studies to identify the forces behind channel selection (e.g., Duguid et al., 1971). Over the past four decades, a number of studies have been published on the determinants and/or outcomes of export channel selection, while currently, there seems to be increasing interest in the topic, with the number of studies has grown in

(12)

2

recent years (Li et al., 2017). By reviewing the current literature on export channel selection, the author finds that the studies in this field: (1) examine various

antecedents to export channel selection; (2) include varied frameworks for the detection of export channel selection, sometimes involving no explicit theoretical foundation; (3) are dominated by one theoretical underpinning – transaction cost analysis (TCA) – and overlook other approaches such as institutions and

resources/capabilities; (4) use quite inconsistent typologies of export channels and descriptions of variables; (5) adopt their own methodology and analysis approaches; and (6) often produce contradictory results, with respect to the influence of

determinants and consequences of export channel strategies. Being such an important aspect of exporting strategy and a decisive route for performance

enhancement, the importance of export channel selection should be highlighted more academically and practically. However, the lack of a systematic effort to

comprehensively examine past work in this field not only limits our understanding of the advancements made in the contemporary literature, but also constrains our ability to explore the new territory in this field. There is, therefore, an urgent need to

synthesise the extant knowledge on export channel selection studies to facilitate theory development and promote advancement in this area.

In addition to performing a systematic review of research into export channels, this thesis finds that there are several gaps that need to be filled in this field to further enrich our understanding of export channel selection.

Firstly, the theoretical bases used in export channel selection need further

development; for example, the applications of resource-based view (RBV) in export channel selection are limited. Although many determinants, including resources,

(13)

3

have been identified as influencing export channel selection, most are based on TCA and regarded as a factor that impacts transaction efficiency (Bello et al., 1995; Klein et al., 1990a; Rialp, 2000). According to RBV, the firm-specific

resources/capabilities owned by firms can play an important role in cost reduction and value creation. In order to realise the value of their resource-based advantages, RBV suggests that firms need to find a strategy to support the exploitation of their resource-based advantages (Barney et al., 2001; Brouthers et al., 2008b; Ketchen et al., 2007). Export channel selection is such strategy that can affect the deployment of firms’ special resources/capabilities. Previous studies of export channels have mainly looked at the impact of firms’ special assets on export channel selection (Li et al., 2017); however, the organisational capabilities that help export firms coordinate export activities and make the best use of basic assets have gone largely overlooked (He et al., 2013; Zou et al., 2003). Although the direct impact of market-oriented capabilities (He et al., 2013) and entrepreneurial-oriented capabilities (Kalinic et al., 2015) on export channel selection has been identified, other capabilities that

highlighted in export research such as product development capabilities (PDC) deserve more attention to enable a full understanding of how firms can use export channels to organise their utilisation of different resources/capabilities. In addition, the impact of firm-specific organisational capabilities on export channel selection is unclear; RBV research demonstrates that the value of firms’ resources/capabilities is not stable and can be conditioned by other factors such as other

resources/capabilities (Carpenter et al., 2001; Shou et al., 2014). This means that the exploitation of particular organisational capabilities can be affected by other

resources/capabilities firms own, and therefore in addition to the direct effect of these capabilities on export channel selection, the moderating effect of certain

(14)

4

capabilities on resource-based channel selection form another serious gap in the literature that deserves research attention. Therefore, it is important to look at the roles of different organisational capabilities in export channel selection for an extension to the application of RBV in export channel selection.

Secondly, more consideration needs to be paid to the application of institutional theory in export channel selection, which examines the influence of institutions on shaping the behaviour and strategic choices of organisations and individuals (Oliver, 1991b, 1997a; Scott, 1995). Institutions have been shown to play an important role in decisions as to whether to enter foreign markets (Peng et al., 2008), and a number of studies of export channel selection provide empirical evidence that institutions exert an important influence on channel selection (Campa et al., 1999; He et al., 2013). However, these studies either fail to follow the widely accepted theoretical frameworks of North (1990) or (e.g., Anderson et al., 1987b; Campa et al., 1999), or mix different components of institutions in one congregated variable (He et al., 2013). According to Scott (1995), the three pillars of institutions – regulative, normative, and cognitive institutions – are totally different, and have their own ways of legitimating firms’ behaviour (Suchman, 1995). Accordingly, the differences in these institutional pillars between home and export markets can have a disparate impact on firms’ channel selection. However, limited attention has also been paid to specific institutional distances, such as cultural-cognitive institutional distance (CCID), which have an important impact on individuals’ and organisations’ strategic decision making (Samaha et al., 2014; Triandis, 1989). Moreover, although RBV claims that firm-specific resources/capabilities can provide sustainable competitive advantages to enhance firms’ performance, some scholars argue that the social contexts in which the resources/capabilities are embedded can affect the value of the

(15)

5

resources/capabilities (He et al., 2013; Peng et al., 2009; Peng et al., 2008). Cultural-cognitive institutions are such a social context; since export firms often operate between two cultural-cognitive institutional environments for confirmatory purposes (Li et al., 2017), the CCID’s influence on firms’ resource-based channel selection is deserving more academic enquiry.

Finally, prior studies into export channels have focused more on the aspect of selection, while the link between export channel selection and its consequence – export performance – has attracted limited attention (e.g., Hessels et al., 2010; Klein et al., 1990a; McNaughton, 1996; Peng et al., 2006). According to the resource-strategy-performance perspective (Barney et al., 2001; Brouthers et al., 2008a; Ketchen et al., 2007; Ray et al., 2004b), firms can select a strategy to support and maximise the value creation of their specific resources/capabilities, and thereby achieve superior performance. Since firms use channel strategies to facilitate the exploitation/enhancement of their firm-specific resources, the strategic fit between their capabilities, institutions, and export channels can significantly influence the result of channel selection and export performance (Barney et al., 2001; Brouthers et al., 2008b). To address this issue, it is important to consider how the alignment between capabilities, conditional factors such as institutional distance, and channel selection affects export performance. However, few studies have addressed this issue in export channel selection (He et al., 2013; Kalinic et al., 2015), and therefore more comprehensive enquiries are required to extend export channel selection research and enrich our knowledge about the performance implications of channel selection.

This thesis fills the above gaps by addressing the following questions: (1) what roles do different organisational capabilities such as PDC and EO play in export channel

(16)

6

selection? (2) How does CCID influence resource-based channel selection? (3) Will firms enjoy a better export performance if the export channels they select are aligned with their capabilities and CCID?

This thesis addresses these questions by using a sample of Chinese export firms to explore the influence of firms’ capabilities (PDC and EO) and CCID on export channel selection and export performance. Based on RBV, this thesis aims to investigate the different roles of organisational capabilities on channel selection and performance enhancement, and also integrate RBV with the institutional theory to explore the moderating effect of CCID on export channel selection, based on differing capabilities. Furthermore, based on the resource-strategy-performance perspective (Barney et al., 2001; Brouthers et al., 2008b; Ketchen et al., 2007), this thesis tests the impact of the fit/alignment between a firm’s organisational

capabilities, institutional distance, and export channel selection on export performance.

By addressing the above issues, this thesis makes the following contributions. Firstly, it attempts to integrate an understanding of export channel selection research by synthesising the existing knowledge, which includes delineating the evolution of the export channel selection literature and the different approaches available to identify the state of the research field. The review conducted here examines, explores, and separates the previous research into theoretical perspectives, antecedents, and outcomes, and considers the data and analytical methodology adopted in these studies to improve our understanding of how the research questions were addressed. Knowledge of the determinants of export channel strategies and their outcomes can be highly beneficial to both academics and practitioners by

(17)

7

facilitating their understanding of the wisdom accumulated in this field, and providing guidelines to help managers make good channel decisions.

Secondly, following the overview, this thesis examines and analyses in detail the various frameworks, theories, and methodologies applied in the research to date from a bird’s-eye view, and thereby appreciate the breadth and depth of current export channel selection research. This comprehensive review contrasts different

perspectives, identifies the most relevant approaches, and specifies the dominant relationships, with the goal of synthesising and integrating the diverse angles

researchers have employed to explore export channel design, and thereby contribute to theory development.

Thirdly, this thesis identifies and discusses in depth a number of important issues in prior research with regard to conceptualisation, theory, and methodology. Based on the discussion, this thesis then recommends directions for further study, such as antecedents and theories that have not been linked to export channel selection, to strengthen the existing theories and frameworks, and possibly touch on the under-debated connections. This thesis also offers suggestions as to how more robust empirical studies could be conducted by considering methodological and statistical issues. It is hoped that these recommendations and suggestions will stimulate further export marketing research on channel strategies and export performance, and thereby develop more theoretical formulations.

Fourthly, this thesis adds RBV to previous TCA-based export channel selection research by adding the organizational capabilities as determinant and moderator of export channel selection. This thesis conceptualises the valuable organisational capability - PDC as a firm-specific resource and uses the mechanism of RBV to

(18)

8

explain how firms can select the appropriate export channel to support the

exploitation of their valuable PDC to improve their value creation from the export channel selection. While previous studies have advanced our knowledge of export channel choice using transaction cost theory (e.g., Campa et al., 1999; Klein et al., 1990a; Peng et al., 2006), they pay large attention to the channel efficiency and focus on the cost-saving effect of the determinant on the transaction cost. However, as channel can not only serve as a way to reduce transaction cost but also a strategy to create value, the differences in firms’ special capabilities and the impact of these capabilities on improving export performance through export channel selection are largely overlooked. By examining the value creation potential of leveraging PDC through export channel selection, this thesis deepens our knowledge of value creation in export markets, i.e., firms can use export channel selection to help them better create value from their PDC and contribute to their performance in export market. Furthermore, this thesis looks at the moderating effect of EO on capabilities-based channel selection; and, by separately exploring the moderating role of EO on PDC-based channel selection, this thesis contributes to the applications of RBV in export channel selection, in that capabilities not only have a direct impact on channel selection, but also moderate channel selections that are based on particular

capabilities.

Fifthly, this thesis contributes to the export channel selection literature by examining the influence of cultural-cognitive institutional distance on firms’ resource-based export channel selection. A number of studies have noted that capabilities that offer competitive advantages in one institutional context may not be valuable in another institutional context (Barney et al., 2011; Oliver, 1997a; Peng et al., 2008). Since the institutional environment varies in every country, differences in institutions will

(19)

9

influence the value of firms’ capabilities, the choice of export channel, and then export performance (e.g., Kalinic et al., 2015; Peng et al., 2008; Yeoh et al., 1995). The cultural-cognitive aspect of an institution is important because it shapes the perceptions, dispositions, and behaviour of individuals/organisations (Samaha et al., 2014; Triandis, 1989). The differences between CCID can result in difficulties in predicting customers’ needs because cultural and cognitive beliefs and value systems in the export market can be very different from those of the home country (Li et al., 2017; Suchman, 1995). As a result, firms’ selection of strategy to support the exploitation of their resources/capabilities will vary according to the CCID between home and export market. This thesis adds to the field by testing and identifying the role of CCID in resource-based channel selections, that is, CCID will impact the PDC-based channel selection through the moderator – EO. While previous studies identify the direct moderating effect of institutional distances such as formal and informal institutional distance on resource-based channel selection (Kalinic et al., 2015), the indirect moderating effect of particular pillar of institutional distance such as CCID is not been identified. This thesis extends previous work by identifying the important indirect moderating effect of CCID on resource-based channel selection through another resource-based moderator, suggesting that CCID will not influence the resource-based channel selection directly but it will influence the channel selection through its impact on particular resource-based moderator. Moreover, this thesis advances the measurement of CCID by looking at managers’ perceived CCID. Since channel decisions are made by managers, the CCID between home and export market varies between export firms. Instead of using the unified distance

measurement from secondary databases such as Hofstede and GLOBE, managers’ perceived CCID can better capture the impact of CCID on firms’ channel selection.

(20)

10

Finally, this thesis provides valuable normative evidence in the export channel domain, which has suffered from a lack of empirical studies that offer performance implications. Due to the fact that performance is rarely discussed in the export channel selection literature, there is a lack of validated and evidence-based support for the assessment and evaluation of the impact of the determinants of export channel decisions (Klein et al., 1990b; McNaughton, 2001; Peng et al., 2006). Previous TCA-based studies focusing on channel efficiency did not provide evidence whether the channel with the cost efficiency effect create better performance. In order to address this issue, this thesis introduces the resource-strategy-performance perspective to explain how the firms can create better performance from export channel selection, that is, firms can select an appropriate strategy that fits their exploitation of their resources to enhance their performance (Barney et al., 2001; Brouthers et al., 2008a; Ketchen et al., 2007; Ray et al., 2004b). By integrating an RBV- and institutional theory-based approach, this thesis finds that firms which align their channel selection with their level of PDC and EO and the CCID between home and export market can achieve better performance. Therefore, this thesis adds to the knowledge by showing that firms can enhance their export performance by using export channel to create value in different cultural-cognitive institutional environment as long as this kind of channel strategy fits the exploitation of their PDC, EO in the export market.

This thesis is organised as follows: Chapter 2 comprehensively reviews and

summaries empirical studies into export channel selection between 1979 and 2015. Then, a discussion of the implications of these studies is presented at the end of Chapter 2. Chapter 3 consists of an empirical study based on this review, exploring PDC as the determinant of export channel selection, and looking at the moderating

(21)

11

effect of EO on PDC-based channel selection. This thesis then explores the three-way interactions between CCID, EO, and PDC in terms of channel selection and the implications of the PDC-based channel selection on export performance. In Chapter 4, this thesis discusses the theoretical and managerial implications of studies into export channel selection.

(22)

12

Chapter 2. A Review of the Empirical Research on Export Channel Selection between 1979 and 2015

Export channel selection is an important strategy for exporting firms. Over the last 45 years, there have been a number of studies investigating the antecedents and outcomes of this strategy. However, no single study systematically reviews the findings in this field. In order to address this gap, this thesis reviews the literature on export channel selection up to 2015 and analyses findings on the determinants and/or consequences of export channel selection. This review shows that in general export channel selection remains underexplored and identifies a number of issues in the current studies, including lacking knowledge of performance implication of channel selection, missing theoretical bases, weaknesses of research methods. Based on this review, the author provides future research directions for development in export channel selection research.

2.1. Scope and analytical approach of the review

In order to undertake a comprehensive search of the studies on export channel selection, this thesis uses advanced search functions, including EBSCO, Science Direct, Scopus, Web of Science, and JSTOR, to identify the export channel selection literature. Keywords related to export channel selection research (e.g., export

channel, intermediary, integrated channel, channel strategy, channel governance, export mode, export integration, export distribution) are used to identify relevant literature without any time restriction. In addition, the author sent out emails on list servers such as the Academy of International Business (AIB) community (one of the largest and most inclusive of its type with over 6,300 subscribers) and asked for studies in the area through their official electronic mailing lists.

(23)

13

Several criteria were established for a study to be included in this research: (1) It must report on a firm(s) that engages in exporting rather than other kinds of foreign market entry modes (e.g., licensing, franchising, joint ventures, or foreign direct investment); (2) it must examine export channel selection from a micro-business perspective rather than that of macroeconomics; (3) it must study export channel selection as a primary and focal objective; (4) it must have an empirical nature which reports data analysis; and (5) it should provide adequate information on research methodologies in order to achieve uniformity and comparability. Case study/research and the literature that appears in non-English publications are not included in this review (e.g., Wen-Shinn et al., 2009). Each article identified by this initial searching process was individually reviewed to ensure that its focal topic was related to export channel selection. Any article that was not topically relevant or did not fit any of the criteria of this review was removed from the sample (e.g., articles focusing on the management of the relationship involved in the export channel rather than selection; articles that study decisions in a given channel rather than the channel selection decision) (e.g., Bello et al., 1985; Chelariu et al., 2006).

After the careful review and selection process, a total of 47 studies were identified (see Table 1), many of which come from leading marketing/international business journals, including International Marketing Review (6), International Business Review (4), Journal of International Business Studies (3), Journal of Marketing (2), Journal of International Marketing (2), Journal of the Academy of Marketing Science (2), Journal of Management (1), Management Science (1), Journal of Marketing Research (1), and European Journal of Marketing (1).

(24)

14 Table 1. Empirical findings of the study reviewed

Author Theoryab _Antecedentse _Mediatorsd _Moderatorsdf Control

Variablesde Dependent Variable (Channel Selection) Performance Related 1 Brady et al. (1979) /

Degree of control (+); Foreign market knowledge (+); Selling cost (mix); Effectiveness of selling product (mix); Amount of export paperwork and document required (+) / / / Direct channel vs. Indirect channel No 2 Anderson et al. (1987a) TCA

Specific assets (+); Product age (NS); Service requirement (NS); Product differentiation (+); Legal restriction (NS); Used channel (+); Relatedness to principal business (NS); Strength of patent (NS); Competitive behaviour (NS); Cultural similarity (+)

/ / / Integrated channels vs.

Independent channel No

3 Klein (1989) TCA

Channel volume (+); Transaction frequency (+); Asset specificity (+); Complexity (+); Dynamism (-)

/ / Share channel (+);

Destination (+)

Degree of vertical

control in export channel No

4 Klein et al. (1990a)

TCA Channel volume (+); Asset specificity

(+); Volatility (mix); Diversity (-); / /

Share channel (+); Destination (+) Hierarchical mode (Foreign) channel) vs. Hierarchical mode (Domestic) channel vs. Intermediate mode channel vs. Market mode channel

(25)

15

Variablesde Dependent Variable (Channel Selection) Performance Related 5 Klein et al. (1990b)

TCA Psychic distance (mix); Export market experience (mix);

Asset

specificity / Share channel (+)

Hierarchical mode (Foreign) channel) vs. Hierarchical mode (Domestic) channel vs. Intermediate mode channel vs. Market mode channel

No

6 Chan (1991) /

Resource availability, Choice of target

market, Firm type / / /

Integrated channel vs.

non-integrated channel Yes

7 Chan (1992) / Home country (NS) / / /

Direct channel vs.

Indirect channel Yes

8 Erramilli et al. (1993)

TCA Asset specificity (NS) /

Capital intensity (mix); Inseparability (+); Cultural distance (NS); Country risk (+); Firm size (+) / Shared-control mode vs. Full-control mode No 9 Grønhaug et al. (1993) /

Firm’s resource base (-); Management education (NS); Manager’s international experience (NS); Product complexity (NS); Market distance and difference (NS); Foreign sales (mix)

/ / /

Company owned subsidiary (operation) vs. Outside agent

(26)

16

Variablesde Dependent Variable (Channel Selection) Performance Related 10 Ramaseshan et al. (1994) TCA

Export experience (NS); Export volume (-); International heritage (-); Age of technology (NS); Profit expectation (NS); Product knowledge (NS); Service requirement (+); Past growth of export market (NS); Potential of export market (NS); Market proximity (NS) / / / Indirect channel vs. Direct channel No 11 Bello et al. (1995) TCA

Specific Assets (+); Volatility (NS); Diversity (-); Export volume (+); Export intensity (+) / / / Non-integrated mode (agent) vs. Non-integrated mode (distributor) No 12 McNaughton (1996)

TCA Channel volume (+); Asset specificity

(+); Volatility (+); Diversity (NS) / / Product Customisation (+); Destination (NS) Hierarchical mode (Foreign or Domestic) channel vs. Intermediate mode channel vs. Market mode channel

No

13 Osborne (1996)

TCA

Specific Assets; Export volume; Firm size; External uncertainty; Product differentiation; Service requirement; Cultural similarity; International experience; Used channel; Political factors / / / Integrated channel vs. Indirect integrated channel vs. non-integrated channel No 14 Aulakh et al. (1997) ET

Asset specificity (mix); Country risk (-); International experience (mix); Firm size (NS); Market position strategy (NS); Global integration strategy (+); Differentiation strategy (mix)

/ / /

Hierarchical mode channel vs. Intermediate mode channel vs. Market mode channel

(27)

17

Variablesde Dependent Variable (Channel Selection) Performance Related 15 Campa et al. (1999) TCA

Intangible assets (+); Product

differentiation (+); Resource availability (+); Export commitment(NS);

Development level of competitor's host country (NS); Potential of export market (+); Institutional and cognitive constraints (+) / / / Internalized channel vs. Shared-control channel No 16 Burgel et al. (2000) SM, TCA, OC

Firm size (+); International experience (NS); Manager’s international experience (NS); Used channel (+); Product technology age (NS); Product customisation (+); Service requirement (NS)

/ / R&D intensity(NS) Intermediary channel vs. Direct channel No

17 Rialp (2000) TCA

Channel volume (NS); Product line (NS); Production technology (mix); Specific Assets (+); Assets technological intensity (mix); Product differentiation (mix); Service requirement (mix); Firm size (+); Resource availability (+); Foreign capital (+); Export commitment (+); Cultural similarity (+); External uncertainty (NS); Foreign distribution advantages (+)

/ / /

Proprietary forms and/or commercial alliances vs. Independent channels

No

18 Kim (2001) TCA, FA

Transaction-specific assets (+); Service requirements (+); Sales value (+); Foreign market experience (NS); Outside distributor’s capability to perform the distribution functions (mix); Age of product (NS)

/ / / Integrated channel vs.

(28)

18

Variablesde Dependent Variable (Channel Selection) Performance Related 19 McNaughton et al. (2001) TCA

Asset specificity (-); Volatility (NS); Diversity (+); Channel volume (-); Product customisation (NS); Destination (NS)

/ / /

Market mode channel vs. Intermediate mode channel vs. Hierarchical mode channel No 20 Chung (2002) /

Firm's characteristics (NS); Product related characteristics (+); Home market position (NS); Potential of export market (NS); Market size of export country (-); Buyers’ business mode (+); Industry difference (NS) / / / Direct channel vs. Indirect channel No 21 Li (2002) REP, TCA

Country-specific knowledge; Superior capabilities; Trust; Market growth; Opportunism; Exporter’s wish to increase coverage

/ / /

Market mode channel vs. Intermediate mode channel vs. Hierarchical mode channel No 22 Li et al. (2002) RCP, TCA, UM

Experiential knowledge (mix), Market turbulence (-); Activity complementarity (+*); Market concentration (NS); Brand power (+); Trust (+*)

/ / /

No

23 McNaughton (2002)

TCA

Asset specificity (-); Volatility (-); Diversity (+); Channel volume (NS); Channel Growth (-); Product customisation (NS); Destination (NS)

/ / / Multiple channel vs.

(29)

19

Variablesde Dependent Variable (Channel Selection) Performance Related 24 Merino et al. (2002) TCA

R&D activities (NS); Standardised product (NS); Level of customer service (+); Service requirement (NS); Brand (NS); Physical and cultural distance (NS); Scale economies effect (+); Number of employee (NS); National ownership (+)

/ / / Proprietary export channel vs. Non-proprietary export channels No 25 Rialp et al. (2002) TCA

Firm size (+); Resource availability (+); Foreign investment (+); Structured planning of export activity (+); Product complexity (+); Product differentiation (+); Promotional activities (+); Level of customer service (+); Industrial Sector (+); Specific foreign market knowledge (+); External uncertainty (+); Export Volume (+); Product line (-); Perception of competitive advantage (+) / / / Proprietary forms vs. Commercial alliances vs. Independent channels No 26 Trabold (2002)

TCA Market Distance (-); Product complexity

(-) / / / Indirect channel vs. Direct channel No 27 Li et al. (2003) TCA, OC, MC

Asset specificity (+); Country risk (NS);

Firm size (+) / / /

(30)

20

Variablesde Dependent Variable (Channel Selection) Performance Related 28 Ekeledo et al. (2004) RBV

Proprietary technology (+); Tacit know-how (NS); Business experience (+); Specialized assets (+); Firm size (+); Organisational culture (+); Company reputation (+); Complementary resource (+);

/ Nature of the

product (mix) /

Sole (Full) control mode vs. Shard control mode No

29 Li (2004) /

Product life cycle; Competition intensity; Differential pricing; Grey marketing; Intermediary power; Broad targeting

/ / / Internet channel vs.

Export intermediatries No

30 Eriksson et al. (2006)

UM

Foreign market knowledge (+); Potential of export market (+); Cultural distance (+); International experience (NS); Customer knowledge (NS); Competitor knowledge (NS)

/ /

Firm size (NS); Firm age (NS); Power distance in the country of origin (+)

Integrated channel vs.

Non-integrated channel No

31 Peng et al. (2006)

TCA Market Distance (-); Product complexity

(-) / / /

Direct export vs. Indirect

export No

32 Lau (2008) TCA

Firm size (+); Firm age (+); Product

complexity (+) / / / Direct channel vs. Indirect channel vs. Multiple channel No 33 Arranz et al. (2009) UM, INVM

Competitive strategy (-); Reactive

strategy (+) / / Industry sector (NS); Firm size (mix); Turnover abroad (mix) Market channel vs. Cooperative channel No

(31)

21

Variablesde Dependent Variable (Channel Selection) Performance Related 34 Carazo et al. (2010) EM

Firm size (+); Firm age (NS); International experience (+); Age of managers (+); Management education (+); Management international experience (-); Specificity of assets for export (+); Specificity of assets for production (+); Transactions frequency (+); Product diversification (+); Foreign market diversification (-); Stimulus in foreign countries (-); Barriers in foreign countries (+); Sector internationalization level (-)

/ / / Direct channel vs.

Indirect channel No

35 Hessels et al. (2010)

RDT, IT

Perceived favourability of home country (mix); Perceived internationalisation of the operation field (NS)

/ /

Industry (+); Firm size (+); Firm age (-); Resource base (NS); Business owner’s education (NS); TMT foreign experience (+); Foreign investors (+) Indirect channel vs. Direct channel No

(32)

22

Variablesde Dependent Variable (Channel Selection) Performance Related 36 Khemakhem (2010) TCA, UM

Product complexity (NS); Service requirement (-); Promotional activities (NS); Product knowledge (NS); Product adaption needs (+); Management goal (-); Management expectation (NS);

Management engagements (NS); Demand condition (NS); Competition condition (NS) / / / Independent channel vs. Integrated channel No 37 Parente et al. (2010) TCA

Cultural distance (NS); Intangible assets (+); Degree of product line concentration (NS); Product complexity (NS); Firm size (+); Advertising intensity (NS); Year-specific effects (NS) Direct writing distribution vs. Independent agency Yes 38 Gabrielsson et al. (2011) UM,

TCA Long-term channel relations / / /

Partner-based channels (indirect, dual, hybrid) vs. Non-partner-based channels (direct) No 39 Abel-Koch (2013) / Firm size (-) / / Firm age (NS); Product innovation (+); Product quality (-); Strength of patent (-); Contract enforceability (NS); Multinational firms (-); Free trade zone (-); Direct import (-); Indirect import (+)

Indirect channel vs.

(33)

23

Variablesde Dependent Variable (Channel Selection) Performance Related 40 Cho et al. (2013) TCA

Searching costs (-); Bargaining costs (-); Monitoring costs (-); Product

standardisation (+); External uncertainty (+); Institutional influence (NS)

/ / / E-intermediary vs.

Market intermediary No

41 He et al. (2013)

RBV, IT Market orientation (+) / Institutional

distance (+); Ownership (mix); Industry (mix); Firm size (NS); Export experience (NS); International experience (NS); Market experience (+); R&D (NS); Frequency (NS); Asset specificity (+); Internal uncertainty (NS); External uncertainty (NS); Market size (NS) Hierarchical channel channel vs. Hybrid (Intermediate) Yes

(34)

24

Variablesde Dependent Variable (Channel Selection) Performance Related 42 Sandberg (2013) NP

Societal Knowledge (+); Business network knowledge (+); Customer-specific knowledge (+) / / / Hierarchical mode (Foreign) channel) vs. Hierarchical mode (Domestic) channel vs. Intermediate mode channel vs. Market mode channel No 43 Fernández-Olmos et al. (2014) TCA, RBV, UM

Firm size (+); Intangible Resources (mix); Product quality (+); International experience (+) / / Business group affiliation (+); Firm age (NS) Direct channel vs. Indirect channel No 44 Dung et al. (2015) PDp

Psychic distance (NS); Entrepreneurs’ age (-); Entrepreneurs’ education (+); Entrepreneurs’ international experience (NS); Entrepreneurs’ social ties (NS)

Entrepreneurs ’ actual behavioural control (NS) Entrepreneurs’ actual behavioural control (NS)

Firm size (+); Firm age (-); Firm’s location (-); Firm’s industry (NS)

Direct channel vs.

(35)

25

Variablesde Dependent Variable (Channel Selection) Performance Related 45 Fernández-Olmos et al. (2015) RBV

R&D intensity (NS); Advertising intensity (NS); Human resources (+); International experience (NS) / / Firm size (+); Information and communication technology (+); Firm age (NS) Direct channel vs.

Indirect channel Yes

46 Kalinic et al. (2015) RBV, IT Entrepreneurial orientation (+) / Regulative institutional distance (-); Normative/cognitiv e institutional distance (mix) Asset specificity (NS); Internal uncertainty (mix); External uncertainty (mix); Frequency (-); Firm size (NS); International experience (NS); Number of countries (NS); Export channel experience (-); Nationality (-); Industry (mix) Hierarchical channel vs. Cooperative channel Yes

47 Serrano et al. (2015)

TCA,

UM The using of Internet (+) / /

Product differentiation (NS); Human capital (+); Firm’s size (+); Firm’s age (NS); Foreign investors (NS) Direct channel vs. Indirect channel No

(36)

26 Notes: a. ‘/’ denotes no theoretical bases have been identified.

b. TCA = Transaction Cost Analysis, RBV = Resources-based View, IT = Institutional Theory, UM = Uppsala Internationalisation Process model, OC = Organisational Capability Perspective, RDT = Resource Dependency Theory, NP = Network Perspective, ET = Eclectic Theory, MC = Marketing Control Theory, SM = Stage Model of

Internationalisation, REP = Relational exchange paradigm, RCP = Relational contracting paradigm, EM = Eclectic model, PDp= Psychic Distance perspective, INVM = International New Venture Model

c. ANOVA = analysis of variance, BA = bivariate analysis, CA = correlation analysis, MDA = multiple discriminant analysis, RA = regression analysis, PA = path analysis, SEM = Structural Equation Model, CTA = content analysis, FA = Functional Approach

d. ‘/’ denotes no mediator/moderator/control variable is used.

e. += increases likelihood of the first channel mode against the rest choices or positive effect on channel internalisation/externalisation, - = decreases likelihood of the first channel mode against the rest choices or negative effect on channel internalisation/externalisation, +*= increases likelihood of the second channel mode against the rest choices or positive effect on channel internalisation/externalisation, -* = decreases likelihood of the second channel mode against the rest choices or negative effect on channel

internalisation/externalisation, mix = mixed result, NS = not significant.

f. += significantly positive impact on the link between antecedent and the channel selection, - = significantly negative impact on the link between antecedent and the channel selection, mix = mixed result, NS = not significant.

(37)

27

This relatively small number of articles on export channel selection is surprising, indicating that this field, whilst having attracted some scholarly enquiries, is not as flourishing as other aspects of exporting such as export performance (Chen et al., 2016; Sousa et al., 2008) and, hence, demands much more research effort to provide richer and robust answers to the focal question of what drives exporting firms’ channel selection. That said, the studies to date have revealed many important

antecedents and consequences of channel selection (e.g., He et al., 2013; Klein et al., 1990b; McNaughton, 2001; Trabold, 2002).

Following the approach used by many scholars on exporting (e.g., Sousa et al., 2008; Tan et al., 2011; Zou et al., 1998), this review employs the vote-counting technique instead of meta-analysis as the analytical method because the latter requires a relatively large sample size (i.e., the number of studies) to establish the relationship between two variables (Hunter et al., 1990), and the articles reviewed cannot meet this specific condition. The vote-counting approach has the advantage that it “summarises for each independent factor, the number of studies that report a

significant positive effect, a significant negative effect or a non-significant effect” on export channel selection, offering a clearer picture for reading (Sousa et al., 2008: 346).

2.2. Theoretical bases and frameworks

In this section, the author discusses the typology of the export channel, and

theoretical frameworks of the studies reviewed. In order to secure a comprehensive view of export channel strategy, the author develops a table that presents the

(38)

28

theoretical bases, analysis method used, and findings of the export channel selection studies included (see Table 1). Due to the complexity of the export channel structure applied in the previous research, this review starts with the typology of the export channel.

2.2.1. Typology of export channel

There seems to be no agreement on a typology of the export channel structure. Hence, there is considerable difficulty in comparing empirical findings. Over 15 typologies are found in previous export channel selection literature (See Table 2). Among them, the direct/indirect channel classification of Brady and Bearden (1979) is the most popular, adopted by 14 studies (e.g., Chung, 2002; Peng et al., 2006; Trabold, 2002). According to them, firms sell their offerings to foreign customers or foreign middlemen/agents/distributors directly or through a company-owned

salesforce/distribution channel located overseas in a direct export channel whereas in indirect channels firms sell to a middleman, agent or distributor who exports for them to the target countries.

Another popular scheme is devised by Klein et al. (1990b). They developed a useful categorisation of three types of channel referring to the market mode, intermediate mode, and hierarchical mode (including integrated channels with offices at home and/or in foreign markets) according to the degree of integration. Ten studies adopt this typology (e.g., He et al., 2013; Rialp et al., 2002). Compared with the

direct/indirect channel typology, the categorisation of

market/intermediate/hierarchical provides a more specific description of firms’ roles and involvement in export activities. Also, the direct/indirect channel typology

(39)

29

includes distributor, agent/middleman in both direct and indirect channel structures, therefore, the differences between these channel members cannot be distinguished clearly. As the role and function of distributor and agent/middleman are quite different in practice, the Klein et al. (1990b) categorisation offer a clearer view of channel structures in exporting.

In addition to these two categorizations, the typology developed by Anderson and Coughlan (1987) which includes integrated and independent channels is adopted by six studies (e.g., Khemakhem, 2010; McNaughton et al., 2001; Ramaseshan et al., 1994).

Service is quite different from other industries due to the specificity and

characteristics of service and its offerings (Kotler et al., 2010). Six studies reviewed look at the channel selection for service industry (e.g., Erramilli et al., 1993;

McNaughton, 1996; Parente et al., 2010). According to the feature of the offering in some non-separable service sector, studies such as Erramilli et al. (1993) and

Ekeledo et al. (2004) developed a classification of shared-control/full control export mode for the channel selection of service firms while direct writing/independent agency typology is used by Parente et al. (2010).

As shown in Table 2, in addition to these typologies, there are a number of

typologies that have only been used once or twice such as proprietary channel/non-proprietary channel classification used by Gabrielsson et al. (2011), single/multiple channel classification used by McNaughton (2002), and hierarchical/cooperative channel classification used by Kalinic et al. (2015).

(40)

30 Table 2. Typology and definition of channel structures

Typology of channel structure Definition of channel structures Studies applied this typology

No of studies applied this typology Direct channel vs. Indirect channel

Firms sell to foreign customers or foreign middlemen/agents/distributors directly or through a company-owned salesforce/distribution channel located overseas

Firms sell to a middleman, agent or distributor who exports for them to the export countries

Brady et al. (1979), Chan (1992), Ramaseshan et al. (1994), Chung (2002), Trabold (2002), Peng et al. (2006), Lau (2008), Carazo et al. (2010), Hessels et al. (2010), Abel-Koch (2013), Fernández-Olmos et al. (2014), Dung et al. (2015), Fernández-Olmos et al. (2015), Serrano et al. (2015) 14 Market channel vs. Intermediate channel vs.

Hierarchical channel (including both domestic and foreign hierarchical mode)

Firms use distributors who take title and perform all marketing and distribution functions Firms use agents or sharing control with another company/agent to perform the

marketing and distribution functions

Firms use the company-owned sales organisation(domestic hierarchical mode)/establish a foreign subsidiary (foreign hierarchical mode) to perform marketing and distribution functions

Klein et al. (1990a), Klein et al. (1990b), McNaughton (1996), Aulakh et al. (1997), McNaughton et al. (2001), Li (2002), Li et al. (2002), Li et al. (2003), He et al. (2013), Sandberg (2013) 10 Integrated channel vs.

Independent channel (or Non-integrated channel)

Firms use primarily captive agents (company salesforce and company distribution division) to perform export activities

Firms use primarily independent intermediaries (outside sales agents and distributor) to perform export activities

Anderson et al. (1987a), Chan (1991), Bello et al. (1995), Kim (2001), Eriksson et al. (2006), Khemakhem (2010),

(41)

31

No of studies applied this typology

Shared-control channel vs.

Full control channel

A share control channel that requires low-to-moderate commitment of resources, exposes the company to low‐to‐moderate business risk, and allows the company

low-to-moderate return on investment

A wholly owned channel enquires the highest commitment of company resources, exposes the company to the highest level of business risk, and allows the highest return on investment

Erramilli et al. (1993), Ekeledo et

al. (2004) 2 Proprietary forms vs. Commercial alliances vs. Independent channels

Firms run commercial facilities abroad on his own

A shared institutional mechanism to develop commercialization and/or distribution activities abroad to take advantage of the partner's physical presence and/or market knowledge of a country-market in question, without the exporter having to establish itself there

Firms carry out international distribution through external intermediaries formed by agents and/or independent distributors in international markets

Rialp (2000), Rialp et al. (2002) 2

Distributor Firms ally with a partner (using a distributor) to perform export activities

Burgel et al. (2000) 1

vs.

Direct export Firms export its offering to foreign market alone (direct exporting)

Export Intermediaries Firms use export merchants or export agents to perform export functions in foreign market

Li (2004) 1

vs.

Internet channel Firms use internet to export to the customer is foreign market directly

Fully internalized channel Firms direct invest in proprietary marketing and distribution abroad

Campa et al. (1999) 1

vs.

Shared-control channel Firms joint ownership of foreign distribution asset or strategic alliances in distribution with firms located in the foreign market to perform export activities

(42)

32

Partner based channel The born global firm in selling to indirect channel partners/or let local distributors become part of a mixed system in which the producer manages numerous customers directly with the Internet, while the local distributors focus on discrete segments of

national markets _{Gabrielsson et al. (2011)} ₁

vs.

Non-partner based channel The born global producer carries out all the channel functions by itself and applies the Internet for both promotion and to generate customers and/or handle product fulfilment Market channel Firms assigning distributors to export

Arranz et al. (2009) 1

vs.

Cooperative channel Firms using cooperation agreements in their exporting activities

Hierarchical channel Exporting firms take full responsibility for distribution and marketing of its products in the foreign country

Kalinic et al. (2015) 1

vs.

Cooperative channel Exporting firms share some of the distribution or marketing with a foreign-based partner through structures such as joint ventures, merchant distributors, and commission agents Market intermediary A specialist firm that functions as the export department of several manufactures in

non-competitive lines to hep firm in exporting

Cho et al. (2013) 1

vs.

E-intermediary An independent market intermediary serving as a B2B electronic marketplace ina form of cyberspace in which qualified members post offers to buy and sell and sales representatives then search the globe for firms that can supply or purchase relevant products, matching exporters with foreign buyers

Integrated channel Firms integrated directly i.e. had set up joint venture or wholly-owned sales subsidiaries without using an existing distributor

Osborne (1996) 1

vs.

Indirect integrated channel Firms integrated through existing distributor vs.

(43)

33

Proprietary channel Firms use vertical integration to perform distributional and sales activities

Merino et al. (2002) 1

vs.

Non-proprietary export channels Firms use external agents or distributors to perform distributional and sales activities Direct writing distribution system A distribution system includes both salespeople employed by the insurance firm and

exclusive agents

Parente et al. (2010) 1

vs.

Independent agency distribution system A distribution system consists of non-exclusive agents

Company owned subsidiary (operation) Firm use the company owned sales operation abroad to handle foreign business activities

Grønhaug et al. (1993) 1

vs.

Outside agent Firms contracting an outside agent to handle foreign business activities Multiple channel Firm use a combination of direct and indirect channel in exporting

McNaughton (2002) 1

vs.

Single channel Firm use a direct or an indirect channel only in exporting Degree of vertical control in export

channel

The degree of centralization and formalization exerted by exporting firms in their export

(44)

34

2.2.2. Theories and frameworks

A number of studies were grounded in different theoretical perspectives, including TCA (e.g., Bello et al., 1995; Klein et al., 1990a), the Uppsala internationalization process model (UM) (e.g., Eriksson et al., 2006; Khemakhem, 2010), the resource-based view (RBV) (including organisational capabilities theories) (He et al., 2013; Li et al., 2003), and institutional theory (IT) (e.g., He et al., 2013; Hessels et al., 2010). Some studies, especially earlier publications, do not explicitly draw on major theories (e.g., Brady et al., 1979; Chan, 1992). This review now analyses the four major theoretical perspectives and the antecedents involved in the studies reviewed (see also Table 1).

Transaction Cost Analysis. Among the theories that have been used, TCA holds a dominant position in explaining export channel decisions, and 29 of the studies reviewed are TCA-based (see Table 1). TCA demonstrates that the decision to apply a particular governance structure depends on the comparative transaction cost (Erramilli et al., 1993; Klein et al., 1990b). Therefore, exporting firms will choose the channel structure that allows them to perform at lower cost, and rely on the market if it is effective (Klein et al., 1990a; Williamson, 1979).

Bounded rationality and opportunism are the two key assumptions in TCA

(Williamson, 1979, 1985). Bounded rationality assumes the constraints of decision makers’ cognitive capabilities and limits on their rationality can become a barrier for firms when facing uncertainties (both environmental and behavioural), which will affect transaction cost (Rindfleisch et al., 1997; Standifird et al., 2000; Williamson, 1979). Opportunism can create problems such as lying, cheating or violating

(45)

35

agreements, and leading people/organisations to behave in their own interests, thus increasing the cost of co-ordination (Rindfleisch et al., 1997; Zhao et al., 2004).

Asset specificity, uncertainty (both internal and external) and frequency are three conditions that are relevant to transaction cost which will affect transaction

arrangements. Asset specificity and internal uncertainty can influence the transaction cost level under the assumption of opportunism, while external uncertainty can influence the transaction cost according to the assumption of bounded rationality (Brouthers et al., 2007; Williamson, 1985). Unlike asset specificity and

uncertainties, frequency is negatively linked to transaction cost as the increased frequency can enable firms to achieve a scale effect that reduces transaction cost (Williamson, 1979, 1985). These three factors in exporting can influence transaction cost levels and, subsequently, export channel arrangements (e.g., Anderson et al., 1987a; Klein et al., 1990a).

Asset specificity refers to the specialised human and physical assets accumulated during the transaction (Klein et al., 1990b; Williamson, 1979). Eighteen studies include asset specificity. Among them, the majority of the studies (15) identify that a high level of asset specificity leads to a greater degree of internalisation of the channel structure (e.g., Klein et al., 1990a; McNaughton, 1996) while three of them fail to have similar results or have mixed results (e.g., Aulakh et al., 1997; Erramilli et al., 1993).

Internal (behaviour) uncertainty arises when firms have difficulty in assessing their partners’ performance under the assumption of bounded rationality (Williamson, 1985). This can be the result of lacking good measures of output or specifying the performance incorrectly (Anderson et al., 1986). Experience is a common way to

(46)

36

measure the internal uncertainty of a firm. When firms have more experience, especially international experience, the internal uncertainty will be lower (Anderson et al., 1986; Brouthers et al., 2007; Zhao et al., 2004). Experience is included in 11 previous export channel studies to explore the influence of internal uncertainty on export channel selection. Although the three studies with significant results all provide support that a higher degree of control in export channel will be chosen when firms gain greater international/exporting experience (e.g., Carazo et al., 2010; Ekeledo et al., 2004), over half of the 11 studies found no evidence or mixed results concerning the connection between internal uncertainty and export channel selection (e.g., Eriksson et al., 2006; Fernández-Olmos et al., 2015).

External uncertainty, or environmental uncertainty, refers to unpredictable changes in circumstances around the exchange (Klein et al., 1990a; Rindfleisch et al., 1997). The unpredictability and changeability of environmental conditions create difficulty for transaction parties in drafting/amending/implementing a contract given their bounded rationality (Klein, 1989; Rindfleisch et al., 1997). Nine studies consider external uncertainty (e.g., Cho et al., 2013; Rialp et al., 2002), five of which examine the influence of the two dimensions of external uncertainty: volatility and diversity, on export channel selection (e.g., Bello et al., 1995; McNaughton, 1996).

Surprisingly, only three studies identify that external uncertainty positively leads to the selection of hierarchical channel/internet as the intermediary/single channel significantly. For the remaining studies, four found mixed effects of volatility and diversity on export channel selection, while two studies found no significant result.

Frequency is used to describe the recurrence of transactions (Brouthers et al., 2007; Williamson, 1985). Often proxied as volume, frequency helps to spread both